主成分回归和岭回归在新疆农业经济的应用  

Application of Xin Jiang Agricultural Economy Based Ridge Regression and Principal Component Regression

作　　者：开璇 张莉莉[1] 

机构地区：[1]新疆财经大学应用数学学院,新疆乌鲁木齐830012

出　　处：《辽宁农业职业技术学院学报》2017年第1期57-61,共5页Journal of Liaoning Agricultural Technical College

摘　　要：基于使用最小二乘回归对数据做统计回归时,常出现自变量之间因为多重共线性,导致模型预测失去效果的现象,其中主成分回归法和岭回归法是常见的用以处理多重共线性问题的方法。现以新疆农业经济数据为例,运用统计软件对数据进行实证分析,证实了这两种方法消除农业经济中多重共线性的可行性,也进一步比较两种方法的实现过程和优缺点,有利于提高回归预测模型的精确性和在现实问题中更广泛的应用。When using least squares regression based on the data for statistical regression, often occurs multicollinearity phenomenon between independent variable and leads model predictions loss effect. Respectively using common methods to deal with the multicollinearity,like principal component analysis and ridge regression.Using statistical software to empirical analysis for the datas of Xinjiang Agricultural economic. Confirmed these common methods are effectively to eliminate multicollinearity,Also learn more about the implementation process and the advantages and disadvantages of two methods, beneficial to improve the accuracy of regression prediction model.and use them in real-world widly.

关 键 词：多重共线性 最小二乘法 主成分回归法 岭回归法 

分 类 号：F372[经济管理—产业经济]